Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A system comprising: a plurality of containers running one or more instances of an application, wherein the application runs on a cluster; and a processor in communication with a memory configured to: scale-down the application until a vitality metric indicates that the application is in a dead state; responsive to the vitality metric indicating that the application is in the dead state, scale-up the application until the vitality metric indicates that the application is in a live state; and responsive to the vitality metric indication transitioning from the dead state to the live state, migrate the application to a different cluster while horizontal scaling of the application is disabled.
2. The system of claim 1 , further comprising an orchestrator with a controller, and wherein the processor is configured to cause the controller to scale-up and scale-down the application.
3. The system of claim 2 , wherein the controller scales-down the application by deleting application instances from a database, and the cluster watches the database to kill an application instance that has been deleted from the database.
4. The system of claim 1 , wherein a binary search is utilized in scaling-down and scaling-up the application to obtain an optimum vitality metric for the application.
A system for optimizing application performance by dynamically adjusting resource allocation to achieve an optimum vitality metric. The vitality metric represents the application's health, efficiency, or performance, and the system scales the application up or down to maximize this metric. The scaling process involves a binary search algorithm to efficiently determine the optimal resource allocation. The binary search iteratively adjusts the application's resource allocation within predefined upper and lower bounds, evaluating the vitality metric at each step to narrow down the optimal configuration. This approach ensures rapid convergence to the best possible performance state while minimizing computational overhead. The system may monitor various application parameters, such as response time, throughput, or energy consumption, to compute the vitality metric. By using a binary search, the system avoids exhaustive testing of all possible configurations, significantly reducing the time and resources required to find the optimal scaling level. The method is applicable to cloud computing environments, virtualized systems, or any scenario where dynamic resource allocation is needed to maintain application performance under varying workloads.
5. The system of claim 1 , wherein scaling-down the application includes reducing a quantity of application instances of the application.
A system for managing application scaling in a computing environment addresses the challenge of efficiently adjusting computational resources to meet varying workload demands. The system monitors performance metrics of an application running on multiple computing nodes and dynamically scales the application up or down based on predefined thresholds. Scaling-down the application involves reducing the number of active application instances to optimize resource utilization and cost. The system may also include mechanisms to distribute workloads across available nodes, ensure high availability, and maintain performance during scaling operations. By automatically adjusting the number of application instances, the system prevents resource over-provisioning while ensuring responsiveness under load. This approach is particularly useful in cloud computing environments where resource allocation must be flexible and cost-effective. The system may further incorporate predictive analytics to anticipate workload changes and preemptively scale resources, enhancing efficiency and user experience.
6. The system of claim 1 , wherein scaling-up the application includes increasing a quantity of application instances of the application.
This invention relates to scalable application deployment in cloud computing environments. The problem addressed is efficiently managing application workloads by dynamically adjusting resources to handle varying demand. The system monitors application performance metrics and automatically scales resources up or down based on predefined thresholds. When scaling up, the system increases the number of application instances to distribute the workload across more servers, improving performance and reliability. Each application instance operates independently but collaborates to handle user requests. The system also includes mechanisms to distribute incoming traffic evenly across instances and manage inter-instance communication. This ensures seamless scaling without service disruption. The invention optimizes resource utilization by adding or removing instances only when necessary, reducing costs while maintaining performance. The solution is particularly useful for applications with unpredictable or fluctuating traffic patterns, such as web services, APIs, or microservices architectures. The system may also integrate with load balancers and orchestration tools to automate deployment and scaling processes.
7. The system of claim 1 , wherein scaling-down and scaling-up the application includes at least one of adjusting disk space allocated to the application, adjusting CPU capacity available to the application, and adjusting memory available to the application.
This invention relates to a system for dynamically scaling computing resources allocated to an application based on demand. The system addresses the problem of inefficient resource utilization in computing environments, where applications may be allocated excessive resources during low-demand periods or insufficient resources during peak demand, leading to performance degradation or wasted capacity. The system monitors application performance metrics and automatically adjusts allocated resources to optimize efficiency. Scaling operations include modifying disk space, CPU capacity, and memory available to the application. By dynamically adjusting these resources, the system ensures that applications receive the necessary computing power while minimizing resource waste. The scaling process may involve increasing (scaling-up) or decreasing (scaling-down) resources in response to real-time demand fluctuations. This approach improves cost efficiency and performance by aligning resource allocation with actual usage patterns. The system may also integrate with cloud computing platforms to leverage elastic resource provisioning capabilities.
8. A method comprising: scaling-down, by a controller, an application until a vitality metric indicates that the application is in a dead state; responsive to the vitality metric indicating that the application is in the dead state, scaling-up, by the controller, the application until the vitality metric indicates that the application is in a live state; and responsive to the vitality metric indication transitioning from the dead state to the live state, migrating the application to a different cluster while horizontal scaling of the application is disabled.
9. The method of claim 8 , wherein the application is one of a plurality of applications in a cluster, the method further comprising: migrating a different application from the plurality of applications in the cluster to the different cluster; responsive to migrating the plurality of applications in the cluster to the different cluster, enabling horizontal scaling for the plurality of applications; and horizontally scaling the plurality of applications.
This invention relates to managing applications in a clustered computing environment, specifically addressing the challenge of efficiently scaling and migrating applications across clusters. The method involves handling a plurality of applications within a cluster, where one application is selected for migration to a different cluster. After migrating the selected application, the system enables horizontal scaling for the applications in the cluster, allowing them to be scaled out or in as needed. The horizontal scaling process then adjusts the number of instances of the applications to meet performance or resource requirements. This approach improves resource utilization and flexibility in distributed computing environments by dynamically redistributing workloads and optimizing application deployment across clusters. The method ensures seamless migration and scaling without disrupting service availability, enhancing system efficiency and adaptability.
10. The method of claim 8 , wherein the controller scales-down the application by deleting application instances from a database.
11. The method of claim 8 , wherein a binary search is utilized in scaling-down and scaling-up the application to obtain an optimum vitality metric for the application.
12. The method of claim 8 , wherein scaling-down the application includes reducing a quantity of application instances of the application.
This invention relates to scaling down applications in a computing environment to optimize resource usage. The problem addressed is the inefficient allocation of computing resources when applications are over-provisioned, leading to wasted resources and increased operational costs. The solution involves dynamically reducing the number of application instances to match actual demand, ensuring optimal resource utilization. The method includes monitoring application performance metrics to detect underutilization. When underutilization is identified, the system scales down the application by reducing the number of active instances. This reduction is based on predefined thresholds or real-time demand analysis. The method ensures that the application remains available while minimizing resource consumption. Additional steps may include gracefully terminating excess instances, redistributing workloads, and adjusting resource allocations to maintain performance. The invention also includes mechanisms to prevent over-scaling, such as maintaining a minimum instance threshold to avoid service disruptions. It may integrate with load balancers and orchestration systems to dynamically adjust instance counts. The approach is applicable to cloud-based, on-premises, or hybrid environments, supporting various application types, including web services, databases, and microservices. The goal is to balance cost efficiency with performance, ensuring resources are allocated proportionally to demand.
13. The method of claim 8 , wherein scaling-up the application includes increasing a quantity of application instances of the application.
This invention relates to scaling applications in a computing environment, specifically addressing the challenge of efficiently managing application workloads by dynamically adjusting resources. The method involves scaling up an application by increasing the number of application instances to handle higher demand. Each application instance represents a separate execution of the application, allowing workload distribution across multiple instances to improve performance and reliability. The scaling process is triggered based on monitored metrics such as CPU usage, memory consumption, or request latency, ensuring resources are allocated proportionally to demand. The method also includes load balancing techniques to distribute incoming requests evenly across the instances, preventing bottlenecks. Additionally, the system may automatically scale down by reducing the number of instances when demand decreases, optimizing resource utilization. This approach enhances system responsiveness and cost-efficiency by dynamically adapting to varying workloads without manual intervention. The invention is particularly useful in cloud computing environments where resource flexibility is critical.
14. The method of claim 8 , wherein scaling-down and scaling-up the application includes at least one of adjusting disk space allocated to the application, adjusting CPU capacity available to the application, and adjusting memory available to the application.
This invention relates to dynamic resource allocation for applications in a computing environment. The problem addressed is the inefficient use of computing resources when applications are statically allocated fixed amounts of disk space, CPU capacity, or memory, leading to either underutilization or resource contention. The solution involves a method for scaling applications up or down based on demand, optimizing resource allocation. The method includes monitoring application performance metrics to determine when scaling is needed. When scaling down, the method reduces the resources allocated to an application, such as decreasing disk space, CPU capacity, or memory. Conversely, when scaling up, the method increases these resources. The adjustments are made dynamically to ensure the application operates efficiently without wasting resources or causing performance degradation. The scaling process may involve multiple resource types, allowing fine-grained control over application performance. This approach improves resource utilization by aligning allocation with actual demand, reducing costs and enhancing system efficiency. The method is particularly useful in cloud computing and virtualized environments where resources are shared among multiple applications. By dynamically adjusting resources, the system avoids over-provisioning and ensures optimal performance for each application.
15. A non-transitory machine-readable medium storing code, which when executed by a processor, is configured to: scale-down an application until a vitality metric indicates that the application is in a dead state; responsive to the vitality metric indicating that the application is in the dead state, scale-up the application until the vitality metric indicates that the application is in a live state; and responsive to the vitality metric indication transitioning from the dead state to the live state, migrate the application to a different cluster while horizontal scaling of the application is disabled.
16. The non-transitory machine-readable medium of claim 15 , wherein the processor is further configured to: migrate a different application from a plurality of applications in a cluster to the different cluster; responsive to migrating the plurality of applications in the cluster to the different cluster, enable horizontal scaling for the plurality of applications; and horizontally scale the plurality of applications.
This invention relates to managing and scaling applications in a clustered computing environment. The problem addressed is the need to efficiently migrate applications between clusters while enabling dynamic scaling to optimize resource utilization. The solution involves a system that migrates applications from one cluster to another and then enables horizontal scaling for the migrated applications. Horizontal scaling refers to adding or removing instances of an application to handle varying workloads. The system first identifies a target cluster for migration, then transfers the applications from the original cluster to the new one. After migration, the system activates horizontal scaling capabilities, allowing the applications to dynamically adjust their instances based on demand. This ensures efficient resource allocation and improved performance. The invention also includes mechanisms to monitor application performance and trigger scaling actions automatically. The overall approach enhances flexibility and scalability in distributed computing environments.
17. The non-transitory machine-readable medium of claim 15 , wherein a binary search is utilized in scaling-down and scaling-up the application to obtain an optimum vitality metric for the application.
This invention relates to optimizing application performance by dynamically scaling resources based on a vitality metric. The problem addressed is efficiently determining the optimal resource allocation for an application to maximize performance while minimizing resource waste. The solution involves a binary search algorithm to iteratively adjust resource allocation (scaling-down or scaling-up) until an optimal vitality metric is achieved. The vitality metric quantifies the application's health or performance, guiding the scaling process. The binary search ensures rapid convergence to the optimal resource level by systematically narrowing down the search space. This approach avoids over-provisioning or under-provisioning resources, improving efficiency and cost-effectiveness. The method is implemented on a non-transitory machine-readable medium, enabling automated and adaptive resource management. The binary search algorithm dynamically adjusts scaling decisions based on real-time feedback from the vitality metric, ensuring continuous optimization. This technique is particularly useful in cloud computing environments where resource allocation must adapt to varying workload demands. The invention provides a systematic and efficient way to balance performance and resource usage, addressing the challenge of dynamic workload management in modern computing systems.
18. The non-transitory machine-readable medium of claim 15 , wherein scaling-down the application includes reducing a quantity of application instances of the application.
19. The non-transitory machine-readable medium of claim 15 , wherein scaling-up the application includes increasing a quantity of application instances of the application.
20. The non-transitory machine-readable medium of claim 15 , wherein scaling-down and scaling-up the application includes at least one of adjusting disk space allocated to the application, adjusting CPU capacity available to the application, and adjusting memory available to the application.
Unknown
March 2, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.